Reliable and Efficient Approximation of Polyconvex Envelopes

From MaRDI portal





An iterative algorithm that approximates the polyconvex envelope \(f^{\text{pc}}\) of a given function \(f: \mathbb{R}^{m\times n}\to\mathbb{R}\), i.e. the largest function below \(f\), which is convex in all minors, is established. The propose iterative algorithm approximates \(f^{\text{pc}}\) of \(f\) as a lower bound of the quasi-convex envelope \(f^{\text{qc}}\). Therefore it is necessary to know upper and lower bounds for \(f^{\text{qc}}\). The aim of the present paper is to establish a reliable and efficient algorithm that computes a lower bound. Numerical schemes for the approximation of upper bounds can be found elsewhere in the literature (the exact citations are given in the references in the end the present paper). A straightforward discretization linearizes nonlinear constraints and results in a large but linear optimization problem. An efficient strategy that reduces the large number of unknowns is proposed. It is shown that for a large class of functions \(f\) the approximation is very accurate. Numerical experiments that illustrate the high efficiency and accuracy of the proposed algorithm are reported.




Cited in
(25)








This page was built for publication: Reliable and Efficient Approximation of Polyconvex Envelopes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5700273)